Adolescence is a period when many individuals begin to experiment with alcohol, with some progressing rapidly to abusive drinking. The acute problems associated with heavy adolescent alcohol use (e.g., alcohol poisoning, sexual victimization, automobile accidents) have been well documented; less certain is whether adolescent alcohol use also has long-term consequences. Three recent reviews of the relevant literature have come to similar conclusions: adolescent drinking is associated with adult functioning, but the existing literature does not tell us whether these associations are causal or whether the consequences of adolescent drinking are long-lasting. We propose to address these limitations with a prospective study of a unique cohort of twins. These twins were initially assessed at age 17, at the early stages of alcohol use, and again at age 20, the period in adolescence when alcohol use peaks. They were subsequently assessed at ages 24 and 29, a period characterized by life-role transition. We now propose to assess them at mid-life (average age of 42 years), a developmental period characterized by stability and for most a moderation of youthful patterns of heavy drinking. The in-person assessment will cover 4 major domains of functioning: 1) mental health, including substance use and abuse, 2) physical health and behavior, 3) neurocognitive function, and 4) psychosocial outcomes including educational and occupational achievement and relationship quality. We estimate that 1130 twins will participate (greater than 90% of the surviving members of the original cohort) in the in-person assessment and that 910 of their spouses/partners will also complete a brief mail survey. Analysis of the data will document the range of adult outcomes associated with adolescent drinking and evaluate three alternative models to account for these effects: 1) the factors that lead to early and heavy drinking in adolescence also increase the likelihood of deficits in adulthood (i.e., confounding), 2) adolescent heavy drinkers tend to become adult heavy drinkers (i.e., drinking persistence), and 3) adolescent drinking upsets the normal course of adult. The innovative cotwin control design, which controls for genetic and shared environmental factors on outcomes, will help isolate possible causal effects of adolescent drinking on midlife functioning. The extensive earlier assessments of this cohort will provide a wealth of information for propensity score estimation to further control for confounding by early indicators of risk (e.g., family history, personality) and provide comprehensive assessments of drinking exposure in early adulthood. Further, we will have an opportunity to investigate whether gender moderates the association of adolescent drinking with adult outcomes. The results of this study will provide essential information about the long-term consequences of adolescent drinking on adult mental and physical health and general psychosocial functioning and determine whether the potentially harmful consequences of adolescent drinking might be ameliorated by drinking desistence or prevented by social factors.